Leak-detection system and methods for leak detection

This case claims priority of U.S. Pat. App. Ser. No. 63/465,475, filed May 10, 2023 and incorporated herein by reference.

The present invention relates to leak detection via infrared cameras, and associated methods.

Early detection of leaks is essential for the safe operation of pipelines, refineries, and other petrochemical installations. To this end, remote sensing of gas leaks in such facilities has been long practiced.

Historically, remote sensing has been performed via satellite imaging or manual observation using infrared-imaging technology. More recently, mobile platforms, such as drones, rail systems, or autonomous vehicles have been used to move thermal imaging and spectroscopy systems throughout large facilities. These mobile systems provide several benefits, including: automation, elimination of redundant systems, and a significant reduction in the overall cost of leak-detection systems.

The imaging of leaks relies heavily on the combination of filters, infrared cameras, and image processing within various spectral ranges. Indeed, imaging with spectrally filtered infrared cameras has emerged as one of the most versatile and efficient tools available for monitoring leaks, as it is possible to detect, and/or differentiate between leaks of methane, propane, and other organic gases.

Presently, filtered infrared cameras are commonly used for periodic inspection, because spectrally selective cameras are very expensive. Such cameras use semiconductor (InSb) detectors, which require cooling, contributing to their high cost. For instance, methane detectors based on semiconductor cameras can cost upwards of $60,000. Recently, infrared cameras based on microbolometer arrays have become available. These cameras are considerably less expensive than the aforementioned cameras; however, their performance is not on-par with that of the expensive cameras.

The present invention provides systems and methods for monitoring/detecting gas leaks, such as from petrochemical, petroleum, or other industrial facilities. In accordance with the present teachings, a leak-detection system employing one or more novel processing methods provides a continuous monitoring system. By virtue of these novel processing methods, relatively inexpensive microbolometer-array-based infrared cameras can be used in the system. That is, notwithstanding a lesser nominal performance specification, the enhancements wrought via the novel processing methods disclosed herein elevate the performance of such IR cameras to levels that meet or exceed those of far more expensive IR cameras. Among any other system benefits, the use of lower-cost cameras enables a large number of cameras to be dedicated to a leak-detection system. Such a system is able to provide more expansive coverage of the monitored facility, as well as other benefits, as described further below.

In some embodiments, the leak-detection system includes a plurality of imaging packages, which include networked IR cameras installed at various locations at an industrial facility. In some other embodiments, the leak-detection system is embodied as a mobile platform, wherein the networked imaging packages are coupled, for example, to flight-capable drones, which fly throughout an industrial facility to monitor for gas leakage. The flight path of the cameras may be pre-programmed and may further be alterable in real time.

The novel processing technique(s) employed by the leak-detection system disclosed herein involve the use of multiple unique filters that enhance the quality (e.g., signal-to-noise ratio, contrast, etc.) of captured images. That is, they enhance the image of a gas leak against other objects. The novel processing techniques include one or more of the following: (i) two-image subtraction; (ii) time-resolved imaging; and (iii) multi-filter detection. In illustrative embodiments, processing techniques (i) and (ii) are used to detect leakage of methane gas, and processing technique (iii) is used to detect leakage of a variety of different gases. Moreover, in embodiments of the leak-detection system as a mobile platform, detection is enhanced by an ability to image leaks from various angles. Additionally, in some embodiments, the imaging system includes a visible (light) camera in addition to an IR camera.

Wherein at least some of the imaging packages include a methane filter that is sensitive (transmissive) to infrared radiation in the range of about 7.7 to 7.8 microns, corresponding to one of the absorption peaks of the IR spectrum of methane, but not sensitive to infrared radiation in the range of about 3.3 microns, which corresponds to another of the absorption peaks of the IR spectrum of methane. Wherein at least some of the imaging packages include a methane filter that comprises alternating layers of germanium and barium fluoride. Wherein at least some of the imaging packages include a methane filter, as well as a reference filter that is substantially not transmissive to infrared radiation in the range of about 7.7 to 7.8 microns; and a processor that subtracts an image obtained from the reference filter from the image obtained from the methane filter. Wherein at least some of the imaging packages include a filter wheel having a methane filter and the reference filter, and a motor that drives the filter wheel to a position one of either the methane filter of the reference filter in front of a lens of the infrared camera, responsive to commands from the processor. Wherein at least some of the imaging packages include a visible (light) camera. Wherein at least some of the imaging packages include a transceiver, separate from the processor, for communicating with a server. Wherein the imaging packages include a power supply for powering elements of the imaging package. Wherein the processor transmits information to the server. Wherein the server transmits information to the processor. Wherein the processor transmits, to the server, information obtained from a visible camera. Wherein the processor transmits data to the server, and wherein the server uses the data to train an A.I. algorithm. Wherein at least some of the imaging packages are coupled to a respective flight-capable drone. Wherein at least some of the flight-capable drones are preprogrammed to monitor a particular location in a facility. Wherein at least some of the flight-capable drones may be directed via the server to monitor a particular location in the facility. Wherein the processor may run an A.I. algorithm that is capable of detecting leaks. Wherein a filter wheel includes a plurality of different filters to monitor a facility for leaks of a plurality of different types of gases. Wherein the processor is capable of subtracting an image obtained via an infrared camera at a first location at a second time, from an image obtained via the infrared camera at the first location at a first time, providing time-resolved imaging. Wherein the time resolved imaging can be of filtered image or a non-filtered image. Wherein images obtained at a first location via an IR camera are obtained at a plurality of angles. Wherein the server redirects the flight-capable drones from pre-programmed locations to another location based on intelligence concerning a suspected leak at the other location. Summarizing, a leak-detection system, and methods therefore, as depicted and described, comprises: (i) a server, and (ii) a plurality of imaging packages, wherein each imaging package includes an infrared camera including a microbolometer array and a processor. Embodiments of the leak-detection system, and methods therefore, may further comprise at least one of the following features, in any (non-conflicting) combination, among other features disclosed herein:

The following disclosure merely illustrates the principles of the invention. It will thus be appreciated that in light of the disclosure, those skilled in the art will be able to devise various arrangements which, although not explicitly described or shown herein, embody the principles of the invention and are included within its spirit and scope. Furthermore, all examples and conditional language recited herein are principally intended for pedagogical purposes to aid the reader in understanding the principles of the invention and the concepts contributed by the inventor(s) to furthering the art, and are to be construed as being without limitation to specifically recited examples and conditions.

Moreover, all statements herein reciting principles, aspects, and embodiments of the disclosure, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents as well as those developed in the future; that is, any elements developed that perform the same function, regardless of structure.

Thus, for example, it will be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative circuitry embodying the principles of the disclosure. Similarly, it will be appreciated that any flow charts, flow diagrams, state transition diagrams, pseudo code, and the like represent various processes that may be substantially represented in computer readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.

1 FIG. 100 100 102 1 102 2 102 3 102 102 101 depicts leak-detection systemfor detecting gas leaks, such as may occur in industrial plants. Leak-detection systeminclude a plurality of imaging packages-,-, and-(hereinafter collectively or individually, imaging package(s)). Imaging packagesare capable of imaging mid infrared radiation (IR), and, in some embodiments, also capable of imaging visible light.

102 100 104 106 104 104 106 In addition to imaging packages, leak-detection systemincludes centralized server, with which the imaging packages exchange data/information/commands over communications links. Servermay be a personal computer (PC), a laptop computer, a tablet computer, or any other computing device. Servertypically includes a processor, main memory, storage device, and input/output (I/O) device, which are interconnected in known fashion. Communication linksmay be supported via Wi-Fi, Bluetooth, or other communications protocols.

100 102 102 100 Systemis depicted as having four imaging packages; this is for simplicity of illustration. It is to be understood that typically, many more such imaging packageswill be included in system.

100 200 200 102 102 1 208 102 102 2 210 2 FIG.A 2 FIG.B Leak-detection systemmay be configured as fixed monitoring systemA, as depicted in, or as mobile monitoring system, as depicted in. In fixed monitoring systemA, some of imaging packages, such as imaging package-, may be installed on tripodat a desired location for leak monitoring in a facility. Some other of imaging packages, such as imaging package-, may be installed on structural supportsor other substantially permanent features of the monitored facility itself.

200 102 212 212 200 200 212 212 102 In mobile monitoring systemB, each imaging packagemay, for example, be mounted on drone. The various dronesmay be pre-programmed to monitor various specific locations in a facility. Mobile monitoring systemB offers a number of advantages over fixed monitoring platformA. For example, one or more dronescan be re-directed to a specific location in real time, such as to investigate a suspected gas leak. Additionally, as described further below, the use of dronescan simplify the technology required in each imaging package.

3 FIG. 102 102 320 322 324 326 328 330 332 334 depicts imaging package. In the illustrative embodiment, imaging packageincludes IR camera, filter wheel, motor, processor, visible camera, power supply, housing, and optional transceiver.

320 In the illustrative embodiment, each IR camerautilizes a microbolometer array as an IR sensor. Well-known in the art, the microbolometer array functions as an uncooled thermal sensor. When infrared radiation having wavelengths between about 7.5 and 14 microns strike the detector, it heats, resulting in a change in its electrical resistance. This resistance change is measured and processed into temperatures that are used to create an image. Such cameras are commercially available from Seek Thermal, Inc. of Santa Barbara, CA and others.

322 322 320 324 322 322 326 320 Filter wheelincludes two or more filters that facilitate the signal-processing methods described later herein. In the illustrative embodiment, filter wheelis positioned in front of the lens (not depicted) of IR camera. Motoris operatively coupled to filter wheelto move a desired one of the filters of filter wheel, per instructions from processor, in position in front of the lens of IR camera.

326 326 102 326 320 326 324 322 326 104 326 104 322 320 Processoris typically a microprocessor, such as a Raspberry Pi brand microprocessor available from the Raspberry Pi Foundation of 37 Hills Rd, Cambridge CB2 1NT, United Kingdom. Processorcontrols/performs a variety of function required for the operation of imaging package. For example, processoris operatively coupled to IR cameraand is capable of activating/deactivating the camera. Processoris also operatively coupled to motor, and is capable of selecting a desired filter of filter wheel. The processor is further capable of performing various signal-processing tasks, such as in conjunction with the methods described later in this specification. Processoris also capable of gathering/organizing data and transmitting it to server. Furthermore, processorcan receive commands from server, such as which filter to select on filter wheel, when to activate/deactivate IR camera.

326 106 104 334 In some embodiments, processorincludes Wi-Fi and Bluetooth capabilities to support communicationsbetween it and server. In scenarios in which Wi-Fi or Bluetooth may not be reliable, or if communications over long distances are required, a separate transceiveris required.

212 104 104 102 It is notable that the use of a mobile platform, such as drones, in conjunction with centralized server, provides an ability to generate large amounts of data with flexible positioning and timing adjustments. For example, leaks may be imaged at various angles to characterize the geometric nature (plume) of the gas leak. Moreover, a mobile platform provides an ability to position other types of detectors and/or imaging systems at a given location for further analysis of the leak, such as gas concentrations within the plume in near real time. In conjunction with server, algorithms may be used to automatically change the performance (e.g., location, timing, etc.) of mobile imaging packagesin response to real-time measurements.

326 326 102 100 104 104 326 102 In some embodiments, processorinclude AI-based algorithms for detecting leaks in real time. For training of the algorithm, processorwithin each imaging packagein systemtransmits real-time video to server. Serverthen trains/updates the algorithm, and updates thereto are then transmitted back to processorwithin each imaging package.

4 FIG. 324 326 −1 −1 A gas, such as methane, may be characterized by its infrared spectrum.depicts the infrared spectrum of methane, shown as a plot of transmittance versus wavenumber. Methane exhibits two absorptions regions: regioncentered at 3.3. microns (wavenumber 3000 cm) and regionat 7.7/7.8 microns (wavenumber 1300 cm). A typical InSb based camera filters the image selectively at the 3.3-micron absorption band, since the bandgap of such a detector array does not permit imaging above about 5 microns. The present inventors preferred to focus on the 7.7/7.8-micron band, since it is more selective in distinguishing between methane and other relevant species.

102 550 5 FIG. Imaging packagemust be outfitted with an appropriate filter so that it is sensitive primarily to the wavelength range of the absorption peak of a gas of interest—such as methane.depicts the structure of optical filterfor detecting methane gas.

550 550 551 552 553 554 555 556 2 t×n/λ t is the physical thickness of the layer in microns; and n is the refractive index of the layer.The optical distance of each layer, and the number of layers in the filter may be adjusted based on the bandwidth and peak position of the transmittance peak. Published Patent Application US2023/0258848, incorporated by reference herein, discloses methods for fabricating optical thin-film structures suitable for use as IR wavelength filters in the range of about 1.5 to 20 microns, as are used herein. Other material combinations, such as Ge and ZnS or ZnSe, or silicon/air, or Ge/polyethylene may also be used as an optical filter for methane, but have been found to exhibit inferior optical properties. wherein: λ is the target wavelength in microns; In the illustrative embodiment, filtercomprises germanium (Ge) and barium fluoride (BaF), which are materials having the highest index contrast and lowest optical loss in the spectral wavelength range from 3 to 20 microns. More particularly, filtercomprises silicon wafer, layerof germanium (optical distance=0.25), layerof barium fluoride (optical distance=0.25), layerof germanium (optical distance=0.51), layerof barium fluoride (optical distance=0.15), and layerof germanium (optical distance=0.35). As used herein, optical distance, which is a dimensionless number, is given by:Optical distance=  [1]

6 FIG.A 6 FIG.B 550 550 642 550 642 −1 depicts a plot of transmittance versus wavenumber for a simulation of filter. The simulation indicates filterprovides peakA at about 7.8 microns (1280 cm) with a FWHM value of 500 nanometers to match the methane absorption line. The measured performance of filter, as depicted in, is consistent with the simulation, showing transmittance peakB at the same location.

102 200 326 A significant challenge in using a fixed camerafor leak detection, as in systemA, is the occurrence of changes in the background. To address this, in some embodiments, a two-image subtraction method is used. In this method, infrared imaging data is obtained using two different optical filters. For example, in the case of methane detection, in some embodiments, a methane filter and a reference filter are used. The images are subtracted from one another, such as via signal processing performed by processor. This results in an enhanced ability to selectively capture leakage of the gas (e.g., methane, etc.) of interest.

7 FIG.A 7 FIG.B 7 FIG.C 7 FIG.D 322 320 depicts the transmittance spectrum of a methane filter, with a transmittance peak at 7.75 microns, anddepicts an image captured using the filter.depicts the transmittance spectrum of a reference filter, as mentioned above, which in this example has a transmittance peak at 6.9 microns, and so captures wavelengths outside the methane-absorption position.depicts an image captured using the reference filter. The methane fume is visible only in the image captured with the methane filter, not the image taken with the reference filter. This proves useful in determining the authenticity of the methane signal. The two filters can be provided on filter wheelfor use with camera.

In some embodiments, two-image subtraction is implemented by obtaining a first image using a methane filter and an IR camera, and obtaining a second image with a visible camera. The visible camera can clarity details in the facility, aiding in pin-pointing the exact position of a gas leak. Note that multiple images of each may be obtained to improve S/N.

In some embodiments, a time-resolved imaging method is employed to maximize methane fume contrast by subtracting one frame from a specific time period from the previous frame. In this method, an image from a specific time period is subtracted from a previous frame to highlight any changes, such as a gas leak. In some embodiments, a methane filter is used for time-resolved imaging. Using the filter, the wavelengths associated with methane absorption can be highlighted. This can enhance the contrast between a methane fume and the background. However, the use of the filter reduces the overall brightness of the image. This may be addressed by measuring without the filter; however, the contrast between the methane emission and the background is reduced in the absence of the methane filter. More generally, data acquired from the time-resolved imaging method, with or without a methane filter/IR camera, may be subtracted from data acquired from the time-resolved imaging method, with a reference filter (a filter that absorbs at a wavelength other than methane, etc.)/IR camera, or an image from the visible camera.

8 FIG.A 8 FIG.B 8 FIG.C 8 FIG.B 8 FIG.A 326 depicts an image taken before a methane leak occurs, anddepicts an image taken after a methane leak occurs.depicts the subtraction of the image offrom the image of, such as performed via process, revealing the leakage.

In a further embodiment, by installing different filters with transmittance spectra at various mid-IR wavelengths, a variety of gases, such as other hydrocarbons (e.g., propane, butane, etc.), as well as greenhouse gases, such as nitrous oxide and chlorofluorocarbons among others, can be monitored for leakage. Such filters can be designed and fabricated in accordance with Published Patent Application US2023/0258848, previously referenced.

9 FIG.A 9 FIG.B −1 320 100 depicts the transmission spectrum of butane, showing an absorption peak located at a wavenumber of approximately 1430 cm(6.9 microns).depicts the transmission spectrum of a butane filter. By installing such a filter in front of IR camera, systemis capable of monitoring for a butane leak.

322 320 Thus, a plurality of such filters, each for monitoring a different gas, can be provided on filter wheelfor use in conjunction with cameras.

200 102 322 324 102 550 320 320 Consider that, in mobile monitoring systemB, imaging packagecan be simplified by dispensing with filter wheeland accompanying motor. In such a mobile system, each imaging packagecan include a single filter. For the two-image subtraction method, for example, a first image can be taken using methane filtervia cameraassociated with a first drone, and a second image can be obtained using the reference filter via cameraassociated with a second drone. Of course, the drones must be capable of precise and rapid positioning.

It is to be understood that the disclosure describes a few embodiments and that many variations of the invention can easily be devised by those skilled in the art after reading this disclosure and that the scope of the present invention is to be determined by the following claims.